1. Field of the Invention
The present invention relates to an image pick-up apparatus, an image pick-up method, and a program for realizing the apparatus and method, which are appropriately used in a digital camera.
2. Description of the Related Art
When a picture is taken with a camera in a long-exposure mode, the camera body often vibrates due to hand shake, which blurs and deteriorates the photographed image. A method for preventing deterioration of the image has been proposed, in which an object is shot continuously at such a high shutter speed that prevents blurring a photographed image, and plural images of the object obtained by the continuous shooting are combined or composed, whereby one sharp image of the object is obtained, as if the object is shot under an appropriate exposure in a normal long-exposure mode.
But even in the above method, if an object to be photographed should move while plural images of the object are being photographed to be combined into one image, the finally obtained image of the object can blur. To solve this defect, Japanese Laid-open Patent Application No. 2001-45359 discloses a technology, in which when plural images of an object photographed during a certain period are combined, an image-to-image variation of locations of the object in the photographed images is detected, and the positions of the images to be composed or combined are adjusted in unit of pixel based on the detected variation. Using the technology, even if the object locates at different positions respectively in photographed images, blur can be avoided in the finally obtained image.
Even though the positions of the images to be composed or combined are adjusted depending on the image-to-image variation of the object locations respectively in the photographed images, a user of a camera gradually changes a photographing or shooting range unintentionally, while he or she holds the camera to shoot the object in a long exposure mode, which causes the following problems.
That is, when the positions of plural images are adjusted in unit of pixel to be combined and composed into one image, the size of such composed image is limited to a size of a common image area among the plural images obtained by continuous shooting. In case that the size of the image to be combined finally is set to a predetermined size, when images to be combined are moved more than a predetermined amount as a shooting range changes gradually, the common image area among these images reduces less than the predetermined size. As the result, an image which is exposed appropriately over the whole image area cannot be obtained. Therefore, to avoid the above defect, the size of the image to be photographed has to be set to a somewhat small size, in other words, a resolution of the photographed image has to be set low. When the image to be composed finally is set variable in size, another problem occurs that the size of an available photographed image reduces (the resolution of the image decreases) as the shooting range changes greatly.
Further, when an amount which a location of an object image changes within image is detected from plural images obtained by continuous shooting, characterizing points used as a basis for combining the images are set in each image and the images are composed or combined into one image based on the characterizing points. In this case, accuracy in detecting the characterizing points decreases as the shooting range changes greatly, and therefore the images cannot be combined precisely. As the result, a quality of the photographed image decreases. On the contrary, to increase accuracy in detecting the characterizing points, still another problem occurs that complex and time consuming processes are required.